Hello everyone, I think I've discovered a bug in JSONDecoder and I'd like to get a quick sanity-check on this, as well as hopefully some ideas to work around the issue. When attempting to decode a Decodable struct from JSON using JSONDecoder, the decoder throws an error when it encounters a dictionary that is keyed by an enum and somehow seems to think that the enum is an Array<Any>. Here's a minimal reproducible example: let jsonString = """ { "variations": { "plural": "tests", "singular": "test" } } """ struct Json: Codable { let variations: [VariationKind: String] } enum VariationKind: String, Codable, Hashable { case plural = "plural" case singular = "singular" } and then the actual decoding: let json = try JSONDecoder().decode( Json.self, from: jsonString.data(using: .utf8)! ) print(json) The expected result would of course be the following: Json( variations: [ VariationKind.plural: "tests", VariationKind.singular: "test" ] ) But the actual result is an error: Swift.DecodingError.typeMismatch( Swift.Array<Any>, Swift.DecodingError.Context( codingPath: [ CodingKeys(stringValue: "variations", intValue: nil) ], debugDescription: "Expected to decode Array<Any> but found a dictionary instead.", underlyingError: nil ) ) So basically, the JSONDecoder tries to decode Swift.Array<Any> but encounters a dictionary (duh), and I have no idea why this is happening. There are literally no arrays anywhere, neither in the actual JSON string, nor in any of the Codable structs. Curiously, if I change the dictionary from [VariationKind: String] to [String: String], everything works perfectly. So something about the enum seems to cause confusion in JSONDecoder. I've tried to fix this by implementing Decodable myself for VariationKind and using a singleValueContainer, but that causes exactly the same error. Am I crazy, or is that a bug?

How to programmatically check if ApplicationToken or ActivityCategoryToken is expired in FamilyActivityPicker? I'm building a Screen Time-based parental control app using FamilyControls and ManagedSettings. We use FamilyActivityPicker to allow the user to select apps and categories to restrict, and we apply the shield using: store.shield.applications = .specific(selection.applicationTokens) store.shield.applicationCategories = .specific(selection.categoryTokens) Sometimes, we observe that the shield silently fails to apply — no error is thrown, but the restrictions aren't enforced. I suspect this may be due to expired or invalid tokens, possibly if the app was removed or the selection became stale. My Questions: Can ApplicationToken or ActivityCategoryToken expire or become invalid over time? If yes, is there a supported or recommended way to detect whether a token is still valid before applying it to the shield? Is comparing the current shield values (store.shield.applications and store.shield.applicationCategories) after applying them a reliable validation method? What's the best practice to handle expired tokens (e.g. re-prompt the FamilyActivityPicker, or show a fallback)? What Is the Expiration Duration of Tokens from FamilyActivityPicker? Any guidance or insight from the Screen Time/FamilyControls team would be greatly appreciated! Thank you!

I have been using Universal Links since January of this year. As of January, it was working fine, but when I checked its operation in August, it was no longer working properly. After investigating, I believe that the reason it is not working is because our firewall is blocking communication from AppleCDN to check for AASA files. Our firewall blocks communication from outside Japan, and Apple's IP address (17.0.0.0/8) is whitelisted. Does anyone know the hostname or IP address that is used to check AASA files? If you know, please let me know.

Hello, How do apps like App Lock implement app-opening from a shield? Is this supported for 3rd-party apps or limited to partners? https://apps.apple.com/us/app/app-lock/id6448239603

Hello everyone, I’m working on an iOS app that uses the new DeviceActivity framework to monitor and report user screen‐time in an extension (DeviceActivityReportExtension). I need to persist my processed screen‐time data into a standalone SQLite database inside the extension, but I’m running into issues opening and writing to the database file. Here’s what I’ve tried so far: import UIKit import DeviceActivity import SQLite3 class DeviceActivityReportExtension: DeviceActivityReportExtension { private var db: OpaquePointer? override func didReceive(_ report: DeviceActivityReport) async { // 1. Construct path in app container: let containerURL = FileManager.default.containerURL(forSecurityApplicationGroupIdentifier: "group.com.mycompany.myapp") let dbURL = containerURL?.appendingPathComponent("ScreenTimeReports.db") // 2. Open database: if sqlite3_open(dbURL?.path, &amp;db) != SQLITE_OK { print("❌ Unable to open database at \(dbURL?.path ?? "unknown path")") return } defer { sqlite3_close(db) } // 3. Create table if needed: let createSQL = """ CREATE TABLE IF NOT EXISTS reports ( id INTEGER PRIMARY KEY AUTOINCREMENT, date TEXT, totalScreenTime DOUBLE ); """ if sqlite3_exec(db, createSQL, nil, nil, nil) != SQLITE_OK { print("❌ Could not create table: \(String(cString: sqlite3_errmsg(db)))") return } // 4. Insert data: let insertSQL = "INSERT INTO reports (date, totalScreenTime) VALUES (?, ?);" var stmt: OpaquePointer? if sqlite3_prepare_v2(db, insertSQL, -1, &amp;stmt, nil) == SQLITE_OK { sqlite3_bind_text(stmt, 1, report.date.description, -1, nil) sqlite3_bind_double(stmt, 2, report.totalActivityDuration) if sqlite3_step(stmt) != SQLITE_DONE { print("❌ Insert failed: \(String(cString: sqlite3_errmsg(db)))") } } sqlite3_finalize(stmt) } } However: Path issues: The extension’s sandbox is separate from the app’s. I’m not sure if I can use the same App Group container, or if there’s a better location for an on‐extension database. Entitlements: I’ve added the App Group (group.com.mycompany.myapp) to both the main app and the extension, but the file never appears, and I still get “unable to open database” errors. My questions are: How do I correctly construct a file URL for an SQLite file in a DeviceActivityReportExtension? Is SQLite the recommended approach here, or is there a more “Apple-approved” pattern for writing data from a DeviceActivity extension? Any sample code snippets, pointers to relevant Apple documentation, or alternative approaches would be greatly appreciated!

Hi everyone, I’m building a sports performance app for Apple Watch that uses the onboard IMU to analyze swings and impacts in sports like tennis and golf. The goal is to estimate club/racket head speed, ball speed, and shot quality in real time from wrist motion data. With Core Motion, I can currently get deviceMotion updates at ~100 Hz. While this is fine for general movement tracking, the actual ball impact happens much faster — 5–10 ms in tennis and ~0.5 ms in golf. Many of the high-frequency vibration/impact components are missed at 100 Hz, making it hard to directly measure or more accurately estimate certain performance metrics. Questions for Apple / community: 1. Is there a way to access raw accelerometer and gyroscope data at higher sampling rates (e.g., 500–1000 Hz) on Apple Watch? 2. If not, is this due to hardware limitations or an API/software constraint? 3. Are there any research, partner, or beta programs that allow deeper sensor access for sports-science use cases? Even modest increases in IMU sampling could unlock more accurate ball-speed estimates, impact force analysis, and strike-quality detection without needing external sensors — making Apple Watch a best-in-class wearable for precision sports analytics. Happy to share more about the current approach, sample data, and potential use cases if helpful. Thanks, Max

I'm trying to work with the beta version of the Declared Age Range framework based on an article's tutorial but am getting the following error: [C:1-3] Error received: Invalidated by remote connection. and AgeRangeService.Error.notAvailable is being thrown on the call to requestAgeRange. I'm using Xcode 26 beta 5 and my simulator is running the 26.0 beta. The iCloud account that I have signed into the simulator has a DOB set as well. This is my full ContentView where I'm trying to accomplish this. struct ContentView: View { @Environment(\.requestAgeRange) var requestAgeRange @State var advancedFeaturesEnabled = false var body: some View { VStack { Button("Advanced Features") {} .disabled(!advancedFeaturesEnabled) } .task { await requestAgeRangeHelper() } } func requestAgeRangeHelper() async { do { let ageRangeResponse = try await requestAgeRange(ageGates: 16) switch ageRangeResponse { case let .sharing(range): if let lowerBound = range.lowerBound, lowerBound >= 16 { advancedFeaturesEnabled = true } case .declinedSharing: break // Handle declined sharing default: break } } catch AgeRangeService.Error.invalidRequest { print("Invalid request") // Handle invalid request (e.g., age range < 2 years) } catch AgeRangeService.Error.notAvailable { print("Not available") // Handle device configuration issues } catch { print("Other") } } }

I'm integrating Twilio Voice (v6.12.1) into my React Native app (using Swift bridging for iOS) and have implemented full VoIP, PushKit, and CallKit support for incoming calls. ✅ What works: Incoming calls trigger the VoIP push and display the full-screen CallKit interface (or fallback UI). Decline Call works as expected. Call logs and events print correctly. ❌ Problem: When I try to accept the call using the CallKit Accept button or React Native fallback UI, I consistently get: ❌ [TwilioVoiceModule] answerCall() callInvite is missing. I also noticed that the console logs for answerCall() are being printed three times in a row, even though the accept button is only pressed once. Additionally in the first time console, answerCall() CallInvite found, accepting... 🔍 Observations: The callInvite is properly received when the VoIP push arrives. But when answerCall() is triggered, callInvite becomes nil. This happens in CallKit accept, RN UI accept, and even notification banner accept. 📦 How it's set up (summarized): PushKit Registration – via PKPushRegistry in AppDelegate.swift VoIP push handling – forwards payload to TwilioVoiceModule.handleIncomingPush() CallKit integration – uses CXProviderDelegate to report and handle accept actions JS Bridge – emits acceptCallAction to JS React Native calls TwilioVoiceModule.answerCall(uuid, callSid) In answerCall() method: self.callInvite is nil – so call cannot be accepted. Attaching the link to view the relevant code: https://docs.google.com/document/d/15pNjKrfk954OaotpMIEh3xQUtst---1K45DBXoYctGM/edit?usp=sharing 💡 Call Accept Flow (iOS): VoIP Push → TwilioVoiceModule.handleIncomingPush() Twilio SDK creates callInvite reportNewIncomingCall() triggers CallKit UI On accept: CXAnswerCallAction → emits event to JS RN calls TwilioVoiceModule.answerCall(uuid, callSid) Problem: callInvite is already nil ❓ What I'm trying to understand: Why is callInvite becoming nil before answerCall() is called? Why are the logs showing the answerCall() call 3 times? Is there a race condition or multiple accept triggers? Should I ensure callInvite is accepted only once across all 3 accept paths? ** 💬 Any help would be appreciated.** 📱 Environment: React Native 0.78.x iOS 17+ Swift bridging with TwilioVoiceModule.swift Twilio Voice SDK 6.12.1 PushKit + CallKit + react-native-callkeep

Hi Everyone, When we first hosted our apple-app-site-association file, our hosting provider was unintentionally blocking Apple’s crawler. As a result, Apple’s CDN seems to have cached a timeout / missing file response. We’ve since corrected the issue — the AASA file is now valid and accessible at: https://our-domain.com/.well-known/apple-app-site-association sidenote: I am using "our-domain" as an alias. It is not our actual domain. We have verified that we return a valid JSON, HTTPS 200, correct MIME type. We used apple recommended tools to check this as well as other tools we found on the internet. However, when fetching through the Apple CDN: https://app-site-association.cdn-apple.com/a/v1/our-domain.com we still receive: Apple-Failure-Reason: SWCERR00301 Timeout Apple-Failure-Details: {"cause":"context deadline exceeded (Client.Timeout exceeded while awaiting headers)"} This has persisted for several days. Tools like getuniversal.link and yURL show that the CDN works fine in U.S. regions, but in Europe it continues serving the old timeout response. I’ve already opened a support ticket (Case ID: 102734912696), but the current support channel seems to be general developer account assistance rather than technical. They claim they can only assist us with account related issues (even though I used the code-support form...) Can someone please advise or help us escalate this to the appropriate internal team to refresh the Apple CDN cache for our domain? Thank you so much for your time and help.

How can we subscribe to over 200 million api calls per month? from WeatherKit api documentation, the max is 200m calls/month

Since iOS 18.4 and macOS 15.4 updating the calendar of an (ek)event (to another accounts calendar) or detaching an repeating event throws an error while saving: "Access denied" EKEventStore.save(event, span: .thisEvent, commit: true) catch { "Access denied" }

Hi, I would like to asking , can I setup a. alarm to alert when phone if OFF power ? since we would like to design a timer with emergence alert. so I need a alert on even phone power is off , Thanks.

Hi Guys, I've been wanting to make a bettermouse-like tool for some time now, but I can't figure out how to modify mouse events to do it. I've managed to get mouse events, like the cursor moving, but I can't seem to modify them. What I try either creates a feedback loop, does absolutely nothing, or makes the cursor have a seizure. Any help is appreciated!

Issue with App Clip Card Showing "App Clip unavailable" for DIGI LIVE App In my DIGI LIVE application, I have an App Clip configured for iOS 17.6+ (the file exceeds the 15 MB limit), and I have made all the necessary preparations for the App Clip to work correctly. However, the App Clip card constantly displays an "App Clip unavailable" error for various users (depending on their system language). The URL associated with the App Clip invocation is: https://ar.digi-live.de The AASA file is located at /.well-known/apple-app-site-association and contains the required fields for the App Clip: { "applinks": { "details": [ { "appIDs": ["N9QR6LF765.de.digilive.app"], "components":[ { "#": "no_universal_links", "exclude": true, "comment": "Matches any URL whose fragment equals no_universal_links and instructs the system not to open it as a universal link" }, { "/": "*", "comment": "Matches any URL like ar.digi-live.de" } ] } ] }, "appclips": { "apps": ["N9QR6LF765.de.digilive.app.Clip"] } } If I use the default App Clip link (https://appclip.apple.com/id?p=de.digilive.app.Clip) to launch it, the "App Clip unavailable" error no longer appears. But if I delete the App Clip from the settings, the error comes back, and it can only be launched again using the default link. It feels like the problem is with the App Clip loading at the moment the App Clip card is triggered — it simply fails to load. Video https://storage.yandexcloud.net/mmrs/files/digilive/IMG_2259.mp4

When an alarm is scheduled for 00:00, the system displays the fullscreen alarm interface twice. After the second fullscreen view appears, both the “Stop” and “Snooze” buttons become unresponsive, but using the physical button (volume or power button) can dismiss the alarm and correctly trigger the StopIntent. This behavior occurs consistently (100%), whether the system time is set normally or manually adjusted before the alarm triggers. Environment: Device: iPhone 14 Pro Max iOS Version: 26.0 Reproducibility: 100%

The name of our app is a portmanteau, and Siri is getting it slightly wrong. It’s putting the emphasis on the second and fourth syllables instead of the first and third. Kind of like if someone starting singing /tinˈeɪʒd muˌtent/. It’s just wrong enough to be funny. It still recognizes the correct name when someone says “hey Siri”. We’ve already tried adding CFBundleSpokenName and INAlternativeAppNamePronunciationHint to info.plist, but neither is changing how Siri says it. We can’t put it in AppIntentVocabulary.plist because we don’t know the key path for the app name.

Hello, currently working on messenger with audio/video calls feature. Have found strange behavior, on some devices after accepting call via Lock Screen video button appears to be disabled. Tested on wwdc Speakerbox sample, same code iPhone 15 Pro (iOS 17.4.1) - video button disabled iPhone 14 (iOS 17.6.1) - video button active Here is my, configuration with supportVideo activated static let providerConfiguration: CXProviderConfiguration = { let configuration = CXProviderConfiguration() configuration.supportsVideo = true configuration.maximumCallGroups = 1 configuration.maximumCallsPerCallGroup = 1 configuration.supportedHandleTypes = [.generic] return configuration }() What can be the reason of this behavior? Thanks

When my Share Extension receives an image from the macOS Photos app on Tahoe, the NSItemProvider passes a URL to an image file in a temporary location. All attempts to read that file fail silently, such as with NSImage(contentsOfFile) I can see that the file does exist in Finder. This code did work in previous macOS versions. It seems like a permissions issue, but startAccessingSecurityScopedResource had no effect. Other platforms work, other apps work, such as shares from Finder, which shares via data instead of a url. I'm really stuck. Has anyone else run into this? // make sure provider has a conforming item if itemProvider.hasItemConformingToTypeIdentifier(imageType) { do { let data = try await itemProvider.loadItem(forTypeIdentifier: imageType, options: nil) // data may be image NativeImage, Data, or a URL to an image file. // figure out which by casting, then attempt to load uiImage if let image = data as? NativeImage { print("found NativeImage") self.images.append(image) } else if let data = data as? Data { print("found Data") if let image = NativeImage(data: data) { self.images.append(image) } } else if let url = data as? URL{ print("found URL") if let image = NativeImage(contentsOfFile: url.path) { print("loaded from URL") self.images.append(image) } } }catch{ print("⛔️ Share Extension Error: \(error.localizedDescription)") } }

I implemented AlarmKit framework to my app. It works fine when the device's screen is off. but it doesn't fire when the screen is on. No sound, No dynamic Island widget. what's wrong... is this just me? I tried other apps on appstore and they worked fine. maybe is it cause they were built with SwiftUI? I'm a flutter developer. just cause my app is non native could this happen... I'm so frustrated.

0 0x102501788 __assert_rtn + 160 1 0x102504570 ld::tool::SymbolTableAtom<x86_64>::classicOrdinalForProxy(ld::Atom const*) (.cold.3) + 0 2 0x10243bdb0 ld::tool::SymbolTableAtom<x86_64>::classicOrdinalForProxy(ld::Atom const*) + 172 3 0x10243cc24 ld::tool::SymbolTableAtom::addImport(ld::Atom const*, ld::tool::StringPoolAtom*) + 140 4 0x10243e508 ld::tool::SymbolTableAtom::encode() + 396 5 0x1024303b0 ___ZN2ld4tool10OutputFile20buildLINKEDITContentERNS_8InternalE_block_invoke.413 + 36 6 0x1886efb2c _dispatch_call_block_and_release + 32 7 0x18870985c _dispatch_client_callout + 16 8 0x1887264cc _dispatch_channel_invoke.cold.5 + 92 9 0x188701fa4 _dispatch_root_queue_drain + 736 10 0x1887025d4 _dispatch_worker_thread2 + 156 11 0x1888a3de4 _pthread_wqthread + 232 A linker snapshot was created at: /tmp/iCSee.debug.dylib-2025-09-18-165226.ld-snapshot ld: Assertion failed: (it != _dylibToOrdinal.end()), function dylibToOrdinal, file OutputFile.cpp, line 5196. clang++: error: linker command failed with exit code 1 (use -v to see invocation)